It is generally desirable in modern radio-telephone communications to maintain a relatively high average carrier modulation level while not exceeding a predetermined maximum modulation level. When a modulated carrier signal is detected in a radio receiver, the output of the receiver detector follows the deviation of the received modulated carrier. It is desirable to modulate the carrier at as high a level as possible (consistent with bandwidth limitations imposed by law and by transmitter and receiver design) to increase the useful receiver output level, and thus increase communications range and decrease degradation of received signal quality due to noise, fading and other factors.
Transmitted carrier modulation level is determined by the amplitude of the intelligence (e.g., voice) signal applied to the modulating circuits (typically, the oscillator or frequency synthesizer in frequency modulation transmitters). Signal processing such as amplitude compression, amplitude limiting and the like can be used to increase the average carrier modulation level. Ideally, average carrier modulation level during useful signal transmission should remain relatively constant at or near the maximum permitted modulation level despite changes in parameters that may affect transmitter modulation level (e.g., carrier frequency, modulating signal characteristics, rf power output, etc.).
Variation in modulation level with change in carrier frequency has not been a major problem in past frequency modulation transceivers because transceivers have generally been incapable of operating outside a relatively narrow operating bandwidth. Hence, most existing FM digital frequency-synthesized transmitters do not provide compensation of carrier modulation level for changes in transmitter operating frequency.
A typical prior-art synthesized transceiver is capable of operating anywhere within a 20 MHz range, and has a maximum carrier deviation level which varies between about 3 and 5 kHz as carrier frequency is changed. In such prior-art transceivers, modulation level (i.e., the effective gain of the modulating circuits) is set to a fixed level such that a predetermined maximum modulation deviation level (e.g., 5 kHz for narrow-band FM) is never exceeded. United States laws governing radio communications do not require a minimum deviation level, but do set a maximum permissible modulation level (and associated envelope bandwidth)--so that adjustment of maximum modulation level is sufficient to guarantee that the transceiver always operates within the modulation level limits set by law no matter what the carrier frequency is set to,
This solution unfortunately has the severe disadvantage that the transceiver carrier modulation level is at or near the maximum permissible level only for a few of the many frequencies at which the transceiver is capable of operating, and is far less than this maximum level at many (if not most) transceiver operating frequencies. Communications range and signal quality and intelligibility are degraded on many transmitter operating frequencies as a result.
FIG. 1 is a graphical illustration of carrier modulation deviation level of an exemplary frequency-modulated radio transmitter plotted against transmit frequency for various methods of modulation deviation level compensation. As the curves marked "A" in FIG. 1 show, variations in modulation deviation level of an uncompensated transmitter can be as much as 3 dB with a change in operating frequency of 20 MHz. To add to this problem, variations in modulation deviation level between different production units may vary by as much as 1 dB or more (as is depicted by three different curves A-1, A-2 and A-3). Thus, the modulation deviation of a completely uncompensated transmitter might fall anywhere within the limits between curves A-1 and A-3. By adjusting the deviation at center band (460 MHz in the example shown) with a variable circuit component such as a potentiometer, the total variation is limited to that indicated by curve A-2 (still 3 dB for a change in operating frequency of .+-.10 MHz).
Some countries in the world regulate minimum as well as maximum FM deviation levels, so that some correction of modulation level with change in transmitter operating frequency is necessary if legal requirements are to be complied with. For example, some transceivers provide a non-linear circuit within the transmitter modulator which adjusts the audio modulation level as a function of synthesizer tuning voltage. This non-linear circuit provides some degree of modulation level correction, but the deviation level is not programmable for individual operating frequencies.
The curve of FIG. 1 marked "B" shows the variation in modulation deviation level for a transmitter having additional non-linear circuit components as described above. Such additional compensation circuit components can reduce variation in modulation level to less than plus or minus 0.5 dB. Unfortunately, unit-to-unit variations in modulation deviation level requires overall modulator circuit gain to be adjusted (e.g., by adjusting a variable resistor controlling the gain of one of the modulator stages) so that the average modulation level is as high as possible consistent with legal requirements. Such adjustments are difficult and time-consuming to make, and must be individually performed for each unit.
A technique which provides constant transmitter modulation deviation level over a wide range of transmit frequencies and which also compensates for variations between transmitter units would increase effective transmission range and ensure compliance with pertinent legal requirements, and would therefore be highly useful.
Another problem which often plagues modern digital mobile radio transceivers is the generation, by the internal circuitry of the transceiver itself, of signals which interfere with reception of desired signals. Sophisticated modern communications devices generally employ high-speed synchronous digital circuitry (e.g., microprocessors) driven by clock synchronizing signals. To obtain the speed performance such digital circuitry is capable of providing, it is necessary to generate one or more clock synchronizing signals of relatively high frequency (e.g., 4 MHz). Harmonics of the clock frequency sometimes fall within the receiver operating frequency band, causing spurious responses which may interfere with the operation of the radio transceiver.
For instance, suppose a radio transceiver capable of operating in the 400 MHz range with a receiver bandwidth of 12 kHz is controlled by an internal microprocessor driven by a crystal oscillator operating at 4 MHz. If the receiver of this transceiver is tuned to an operating frequency of 464.000 MHz, a clear-channel spectrum of 463.994 MHz to 464.006 MHz must be maintained to avoid interference with signals to be received. Unfortunately, an on-channel spurious response may be caused by the 116th harmonic of the microprocessor oscillator (4.0 MHz.times.116=464.000 MHz).
One solution to the problem described above is to significantly improve the shielding between the radio frequency circuitry and the digital circuitry of the transceiver so that no signals generated by the digital circuitry can find their way into the RF circuitry. Because of the high sensitivity and the compactness of modern radio transceivers, it is extremely difficult to provide sufficient shielding to guarantee complete freedom from receiver interference.
Another method sometimes used to reduce interference of received signals caused by spurious signals generated by a microprocessor clock oscillator is to select an oscillator frequency which is not harmonically related to any frequencies desired to be received. Although this solution works well for transceivers having only a few fixed operating frequencies, modern microprocessor-controlled radio transceivers are capable of operating on a large number of channels and can be programmed (or reprogrammed) in the field to operate on new or different channels than those selected at time of production. It is sometimes impossible to locate a clock oscillator frequency which is not harmonically related to any of a large number of possible operating frequencies. Moreover, the clock oscillator frequency may have to be changed every time the transceiver is programmed to operate on new or different operating frequencies, thereby increasing field servicing time, cost and complexity. An arrangement which guarantees complete freedom from internal oscillator-generated receiver interference regardless of receive frequency would be very valuable and useful.